We used population sample data from 25 societies to answer the following questions: (a) How consistently across societies do adolescents report more problems than their parents report about them? (b) Do levels of parent-adolescent agreement vary among societies for different kinds of problems? (c) How well do parents and adolescents in different societies agree on problem item ratings? (d) How much do parent-adolescent dyads within each society vary in agreement on item ratings? (e) How well do parent-adolescent dyads within each society agree on the adolescent's deviance status? We used five methods to test cross-informant agreement for ratings obtained from 27,861 adolescents ages 11 to 18 and their parents. Youth Self-Report (YSR) mean scores were significantly higher than Child Behavior Checklist (CBCL) mean scores for all problem scales in almost all societies, but the magnitude of the YSR-CBCL discrepancy varied across societies. Cross-informant correlations for problem scale scores varied more across societies than across types of problems. Across societies, parents and adolescents tended to rate the same items as low, medium, or high, but within-dyad parent-adolescent item agreement varied widely in every society. In all societies, both parental noncorroboration of self-reported deviance and adolescent noncorroboration of parent-reported deviance were common. Results indicated many multicultural consistencies but also some important differences in parent-adolescent cross-informant agreement. Our findings provide valuable normative baselines against which to compare multicultural findings for clinical samples.

As the US population becomes more diverse in the 21st century, researchers face many conceptual and methodological challenges in working with diverse populations. We discuss these issues for racially and ethnically diverse youth, using Spencer's phenomenological variant of ecological systems theory (PVEST) as a guiding framework. We present a brief historical background and discuss recurring conceptual flaws in research on diverse youth, presenting PVEST as a corrective to these flaws. We highlight the interaction of race, culture, socioeconomic status, and various contexts of development with identity formation and other salient developmental processes. Challenges in research design and interpretation of data are also covered with regard to both assessment of contexts and developmental processes. We draw upon examples from neighborhood assessments, ethnic identity development, and attachment research to illustrate conceptual and methodological challenges, and we discuss strategies to address these challenges. The policy implications of our analysis are also considered.

No abstract